Computer ProgrammingLanguages are only a subset of the different languages found on Earth. All languages from the most simplistic odors and back-archings to the most complicated multi-lingual pattern labyrinths are used to convey information and direct the actions of others. In order to be effective at communicating and directing actions in a complex world, many different languages must be employed. As soon as one's baby brain realizes how to use sounds and body language to get someone to do something, one becomes a programmer. While a child is a considerable novice to programming, an effective politician or entertainer is a master programmer. It's as simple as getting food and as complex as controlling the populations of several countries.
"Whenever you get feedback from a computer, it's not the results of your programming - it's the computer programming you. AllLanguageIsProgramming InSovietRussia."
What you say may be true, but if in this forum you insist on using the term "programming" to refer to something other than "computer programming" then you risk misunderstanding. Are you being misleading intentionally?
- Programming is the superset of computer programming. There is no trickery meant. I meant what I wrote.
- So then this implies that dolphins and ants program?
- Clearly - he means that everyone and everything that successfully communicates anything and thereby causes a desired reaction "programs". Interesting use of the word.
- I'm glad you agree. I can also 'program' the dirt to give forth a tree in the same way I can 'program' a computer to give forth a directory map.
- He didn't mention intentionality nor distinguish teleonomy from teleology, so evolution must program too. And gravity programs.
- Gravity is language... evolution is language... carrots are language... ahmmmmm.... ahmmmm... ummm....? That's an interesting take on those theories and/or laws. I'm damned for not having thought of it before!
- Carrots are language, so language improves your eyesight. Corollary: Bugs Bunny was a programmer.
- Therefore, Elmer Fudd, Marvin The Martian, and Yosimite Sam were attempting to debug. De-Bugs?
AllLanguageIsProgramming, so therefore I can implement the 8 queens problem in HTML. Better yet, I can do it in wiki markup. Let's see, was that 19 single quotes or 20....
Yeah? So let's see you do it.
I asked you first.
No you didn't!
I know you are, but what am I?
[What's that got to do with 8 queens?]
You have to pick the language first. But under the AllLanguageIsProgramming hypothesis, picking the language is undecideable. So anything can be programmed in theory, but nothing can be in practice.
- Different languages have different capabilities, but that doesn't keep them from adhering to 'all language is programming'. Markup language is a subset of language and programming languages and therefore doesn't define either.
OK, so it is an unusual, perhaps counter-intuitive use of the word, but that doesn't mean that the concept is without merit. Now that we've had our fun making fun of it, are there lessons, parallels, or connections to find and learn from?
- The concept has merit, but to keep things straight, why don't we call it by the more usual word communication and reserve the word programming for what it normally means? Then we can take advantage of what people already know about communicating, without confusing them.
- Then it would be 'all language is communication' and we would reduce the entire statement to elementary school logic, which absolutely isn't the point. Programming is language, but the point missed is language is programming. If language didn't convey information meant to control something, then there wouldn't be a reason to have language in the first place. In fact, if there weren't any language - of any sort - then there wouldn't be a need for senses. Or thought. Or existence. Programming produces cause-and-effect. So does language. Language is an attempt to produce a given cause-and-effect and therefore is a kind of programming (the word 'programming' wasn't originally attached to the word 'language', nor vice-versa).
Except that's still elementary school logic, but now it's only true if you extend the word
programming beyond its ordinary meaning. Why that speaking with bent language of that sort should be considered more profound than making the same statement using the ordinary terms is beyond me.
Programming (n.),
- 1. Creating a sequence of instructions to enable a computer to do something. (In this case, the computer can be a human brain.)
- 2. A pastime similar to banging one's head against a wall, but with fewer opportunities for reward. (In this case, my head versus your wall.)
We're attempting to use language to program each other into an arrangement of thought. Programming a PC isn't much different. One is hardware (the PC) and the other is
WetWare (the human brain).
However, another interesting point is: by the vary nature of language itself, language attempts to program.
JMN, I like the concept of language as programming. It seems to me that talking to someone can be thought of as building 'state' in the recipients mind in a manner analogous to setting variables; albeit in an inexact and fuzzy manner. Consider the act of telling a joke; it requires certain preparatory statements to have been entered to build up a particular state so that the punchline will perform the desired action - presumably laughter/amusement. The difference between this and plain 'communication' is that communication implies a mere data transfer - whereas the idea of language as 'programming' suggests the speaker/author has a concept of the sorts of mental state that their words can build up, and crafts their delivery to tend towards certain 'results'. Whether anyone can actually do this with enough precision to really be considered programming is debatable; for example, programming girls with cheesy pickup-lines is kinda like feeding python statements into a perl interpreter ;)
oops - there's a bug in the punchline. s/cheesy/suggestive/
Spoken language used to psychologically manipulate people as if they were inanimate objects is widely considered unethical, so if that is the main value of the thought behind AllLanguageIsProgramming, then there isn't very much value there.
Why would the fact that people think it's unethical (assuming they do) nullify the value of this idea?
That's a good question. I guess I was assuming there's no point in exploring an unethical approach, but that's obviously wrong to the extent that there can be value in describing things without any intent to practice them.
On the other hand, there's still the critique that this implies (as you said, "assuming they do [think it's unethical]") that AllLanguageIsProgramming has inherent ethical issues, while conventional computer programming does not (there are many situations where ethics arises, but they are dependent on final use, not on the act of programming the computer in and of itself).
So there's at least one sharp difference, despite any claimed similarities.
But if all language is programming (in some sense, from some perspective), should we avoid discussing that because it's widely considered unethical? Being unethical has no bearing on its truth. What if we can't help but use language to manipulate people as if they were inanimate objects? Or even animate objects? I'm looking at this from the perspective of Dennett's model of consciousness as internalized speech. In that model language evolved to program/control people (not just other people, but ourselves as well). If that's unethical then language itself is unethical, because it serves no other purpose than to attempt to change the mental state (and consequently the actions) of a human.
Heh. Yeah, well, Dennett is a very smart guy, but he is definitely prone to counting angels dancing on pin heads. You hadn't noticed that the word "manipulation" or "manipulative" has a very negative connotation?
I don't deny the negative connotation, only question why that connotation would decrease the value of this page. If language is programming us (and I believe it is) there's value in examining how it does that. And I don't understand what his model of consciousness and its evolution has to do with angel counting.
- It's a classic way to accuse philosophers of MentalMasturbation. Dennett is brilliant, and well known for interesting writings about things like consciousness. Is he also authoritative about consciousness? Of course not; no one is. It's still an ill-defined term, let alone well-understood. So what does his model have to do with anything at all? Nothing; it's not clear to me that he has something that is specific enough to call a model.
- Indeed, he calls it a "just-so story", but I find it convincing.
The difference is that once people are conscious, one can distinguish between voluntary and involuntary changes. Manipulation and programming nearly always refer to involuntary changes, which are the only kind available on a computer.
Persuasion is something different; while it might have the same sort of goals, it functions very differently in practice. But assuming Dennett's hypothesis, or even something like Jaynes, it might be fair to say that language started out as programming.
- No one assumes Jaynes. Not many do, but I've seen it done, both by people and by literature. A weakened version fits well with some other hypotheses, Dennett's included, which is why I mentioned it.
But consciousness doesn't give us the ability to avoid involuntary manipulation. If you see a gruesome image of a murder victim on the news, you may close your eyes or change the channel (voluntarily) but your brain may reproduce the image for some time afterward (involuntarily). The same is true of language, even more so in some ways, because you can't close your ears.
- It's cheating to say "involuntary" there; that sort of makes it a tautology. Try it this way: does consciousness give us the ability to avoid any kind of manipulation? Sure, obviously. In fact, detecting deception and manipulation is one of the evolutionary advantages of consciousness.
- But consciousness also enables deception and manipulation, and not just in the deceiver/manipulator. Try convincing your cat that Jews control the world's money supply, or that things go better with Coke. Mine isn't buying it. These kinds of programming are only possible if the listener is conscious. Consciousness can allow us to direct some of our own programming and filter the programming of others, but it doesn't protect us from "involuntary changes" as argued above.
- You're confusing 'consciousness' with 'understanding'. A cat doesn't understand such things, but if living and coherent to the outside world, is quite conscious of your strange behavior and sounds.
- [agreed]
- We're using different definitions of consciousness, then. I don't mean awake, I mean "[a] sense of one's personal or collective identity, including the attitudes, beliefs, and sensitivities held by or considered characteristic of an individual or group". I thought that's what we were talking about.
- Cats have all of those capabilities, it's just that some of them are sharply limited in degree by their limited capacity for understanding. Thus there continues to be an issue with conflating consciousness and understanding.
Of course, but not
all language falls into this category. As with any stimulus there will always be some involuntary effects, but they may not be an dominant or even notable part of the response. And when they are, they don't always correlate with what the speaker intends, making their control very limited.
Can we think of some use of language that doesn't fall into this category? I agree that the effect isn't always what the speaker intended, but that's true of computer programming as well. When I look at the effect a well constructed pamphlet, theorem, poem, song, article, speech or advertisement, the control doesn't seem "very limited".
Sometimes you can notice you are being influenced and resist. You may notice someone is obviously trying to make you mad, and you consciously refuse. Same with advertising.
Sometimes, yes, but if language wasn't an effective way to program people, why would there be so much advertising?
Language is effective, but only some language attempts to influence in this manner. For instance, people bounce ideas off of each other, looking to gain insight rather than change the other person's state of mind - or their own, at least not towards any specific end. And while programmers make mistakes, they can hardly be compared to the completely unpredictable results of some dialogues. If you'll notice, I said the difference was a matter of degree; while there are some similarities, in the vast majority of cases they aren't enough to make the two apply in similar ways. Good advertising may be an exception, but it's a long way from AllLanguageIsProgramming.
- This reminds me of the different kinds of speech, e.g. statement, question, command, speech-act.
"Bouncing ideas" off of other people is intended to influence them. It's a simulation or test of how those ideas affect others. What kinds of language don't attempt to influence people?
- Depends on how you mean "influence". In a sense similar to "program/manipulate people", we've mentioned some exceptions already. In the very widest and loosest of senses, there is some argument that could be made that all kinds attempt to influence people one way or another (not an airtight argument, but a reasonable one). That seems different from the page topic, though.
- Note the above about different kinds of speech. Many programmers would say that their programming is always a matter of commands, and that they never do queries/questions (I'm pretty sure I heard someone say that recently :-) And we then see trivially that only 1/4 of those language types (statement, question, command, speech-act) are programming. Or does "influence" also mean "question" sometimes? I wouldn't think so.
- If I ask you how old you are and you respond, I've altered your behavior. You wouldn't have stated your age (or lied) without me asking the question. A question is a very direct kind of programming.
- I'm going to tell Top that you believe queries are programming. :-)
- Altered behavior, yes, that's true, but I don't think that this is the same as treating someone like an inanimate object. We rely on people's good will to answer our questions; it is usually a matter of their consciousness and free will.
- Yes, we treat them as animate objects, and we don't normally associate the verb "program" with animate objects, but that's because computers are still so dumb. When they are as "conscious" as we are, will we still use the verb "program" to describe the act of changing their behavior?
- If they have free will and sometimes tell us "I don't feel like it", then no, that won't be programming.
- How free can their will be if we can change their code? I think this is the crux of the issue. I don't believe in free will because it is possible to change the configuration of the human nervous system and alter the way a human thinks and acts. I see language as one useful (if imprecise) tool for doing that.
- Yes, but that doesn't mean that free will never exists, it just means that it may not always exist and/or can be diminished in degree. Anecdotally, it seems obvious to me by observation that some people operate as automatons without anything much resembling free will, but I don't think they are representative of everyone.
- I hope you realize that free will is one of the classic unsolved problems on which many disagree, so let's not digress too far in that direction.
- I think our differing views on free will inform our views of language as programming. I don't see how free will can ever exist if everything I think or do is determined by the laws of physics. If you could reach into my brain and change its structure, you could make me think I had free will. You could even make me think I came to that conclusion as a result of my free will. And right now I'm trying to reach inside your brain and make you think you don't. I don't expect to succeed, but that is my intent.
- For starters, physics is not deterministic; that's a pre-20th century kind of physics. If you know the precise state of the entire universe in one moment, it is not possible even in principle to use that information to predict future states of the universe.
- Reductionism nonetheless does seem to offer only a limited number of choices, that humans are governed by one of (or a mixture of) either deterministic mechanism, randomness, or some kind of metaphysical dualism...however...
- Reductionism, although obviously useful, has long-established limits. Chemistry theoretically derives from the laws of physics, but in practice doing so is vastly beyond our understanding (with certain limited exceptions). Same thing but more so in attempting to reduce biology to chemistry. Reductionism fails in many places due to emergent properties of complex system; chaos theory shows that you can never have enough computational power to fully predict those emergent properties.
- Free will has never been successfully attacked with reductionism, but that's not surprising given the above, nor does it mean it doesn't exist. It's an emergent property of an extremely complex system. It furthermore has been demonstrated in multiple domains (ethics, law, politics, many areas of psychology) that it is highly useful to recognize free will as something very real, and the failure of reductionism to encompass free will as an outgrowth of principles of lower level scientific systems means nothing at all.
- To put it another way, although some people philosophically choose to believe there is no free will, nonetheless we are all obligated to act as if it free will does exist. The police and courts, among others, will not be amused by arguments to the contrary.
- Are you trying to reprogram me with all of this language?
- :-) I think that's a perfect note on which to let you have the last word.
[Also, as was suggested thousands of years before Jaynes, not everyone is conscious when they're awake (ever drive home on automatic while daydreaming?) - such awake but semi-conscious people can be highly influenced by hypnogogic suggestions such as advertising.]
Would someone now care to re-write the above as DocumentMode?
- ThreadMode can be used to lead-up to decent document-mode statements. Let it play out and there may be some good material to convert to document-mode, or perhaps something will spark a good document-mode statement from someone.
- Also, this suggestion was made when there were only about 6 comments on the page, 5 of which were jokes, so it was absurdly premature.
CategoryNaturalLanguage CategoryProgrammingLanguage
All Language Is Programming